Glossary · Term

bf16

← all terms

Definition

Plain language

A compact way of storing the numbers inside a neural network, using fewer digits to save memory and run faster.

As stated in the literature

bfloat16, a 16-bit floating-point format with an 8-bit exponent; its limited precision sets the rounding-error floor that interpretability work must exceed to claim an intervention is real signal rather than noise.

Also called: bfloat16

Why it matters: Its limited precision sets a floor on rounding error, so interpretability claims have to clear that floor to prove an effect is real and not just numerical noise.

For example, storing a weight as bf16 keeps roughly the right size of the number but drops some of its trailing precision to save memory.

Heard on the show

“Right — a skeptic would worry the iteration deltas are just bf16 rounding noise.”
Episode 032 — A Sticky-Note for Every Layer: Letting Transformers Remember What They Were Just Thinking

Mentioned in 1 episode

  1. 032
    A Sticky-Note for Every Layer: Letting Transformers Remember What They Were Just Thinking

Related terms